Ideating a Recommender System for Business Growth Using Profit Pattern Mining and Uncertainty Theory

Abstract

Association rule mining (ARM) alone is a classical yet powerful method for basic rule discovery. However, generic measures being used are insufficient for specific pattern generation and rules of business interest. Critical decision making is a “key” component in contemporary businesses which could be rewarded by periodically utilizing patterns and rules to steer business growth and profit as well. To effectuate self-propelled growth in businesses, a feasible optimal recommender system needs to be accomplished without human intervention that recommends targeted product marketing and promotional strategies. In conjunction to ARM, uncertainty is a growing challenge in data mining research with facets of being probabilistic, fuzzy, or vague. Among many set theories to surmount uncertainty, vague set theory is employed for handling vagueness in data which gives the motivation of implementing a knowledge-based recommender framework by aggregating the two approaches to predict uncertain market growth strategy patterns and profitable rules.